class: center, middle, inverse, title-slide # Segregation and Affirmative Action ## Evidence from colleges in India ### Saurabh Khanna ### Stanford University ### 2020-11-18 --- # Research Questions 1. How do institutional policies affect segregation based on affirmative action quotas? -- 2. Do these effects differ for advantaged versus disadvantaged students? -- 3. Do these effects differ at elite versus non-elite institutions? -- <br/><br/><br/> Why? --- class: inverse, center, middle # Data --- # Data * 50 randomly sampled colleges in India + 200 classrooms (4 per college) - ~20000 students (averaging 100 per classroom) -- * Network data for each classroom + Students nominate upto 10 friends + Longitudinal data - 2 time points -- * Affirmative Action Policy + Quotas _reserved_ for students from historically disadvantaged backgrounds + __Reservation students__: Those who gain admission based on these quotas + __Non-Reservation students__: Those who don't --- class: inverse, center, middle # Method 1 ### Regression on Segregation Scores --- # Segregation Score <table> <thead> <tr> <th style="text-align:left;"> </th> <th style="text-align:left;"> Friend (f) </th> <th style="text-align:left;"> Not friend (f') </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> Reservation (r) </td> <td style="text-align:left;"> A </td> <td style="text-align:left;"> C </td> </tr> <tr> <td style="text-align:left;"> Non-reservation (nr) </td> <td style="text-align:left;"> B </td> <td style="text-align:left;"> D </td> </tr> </tbody> </table> <br/> -- Natural log of the odds that a student will nominate a friend in their own category, relative to the odds that they will nominate someone in a different category `$$segregation_{r} = log \left[\frac{f_{r}}{f_{nr}}\times\frac{f'_{nr}}{f'_{r}} \right] = log \left(\frac{AD}{BC} \right)$$` `$$segregation_{nr} = log \left[\frac{f_{nr}}{f_{r}}\times\frac{f'_{r}}{f'_{nr}} \right] = log \left(\frac{BC}{AD} \right)$$` <br/> -- `\(+\)` score `\(\to\)` More within category friendships <br/> `\(-\)` score `\(\to\)` More across category friendships <br/> 0 `\(\to\)` Optimum within/across category friendships (Preference = Opportunity) ---